Can you convert XY Vector to a LCD?

[TheBman80]

The reason I’m asking this question is that I stumbled across a sight that can convert Mame to a vector monitor for old school vector games. http://www.zektor.com/zvg/ This is all well and good if you want to use your computer as the arcade platform and you just happen to have an extra working COLOR Vector monitor lying around just collecting dust. But what if you don’t have a Vector monitor and have a full Vector game without one and NEED an easier solution then spending a small fortune to get a Vector monitor. Then there’s always the risk that the monitor will crap out on you after you get it lets face it there old and sometimes they wear out. Has anyone heard of some electronic component that you could wire the PCB or use the original wiring harness and just plug it into a LCD and never having to worry about monitor failure again. Oh and I’m sorry to those I offended by suggesting putting Vector games on to a nonvector monitor. This would also be a great solution to those people who have vector games without the necessary monitor to actually get some use out of their cabinets other than converting them to something else.

[danwinslow]

Well...its not the monitor that controls how the graphics gets generated, it's the computer. The difference between a raster output and a vector output is not just one of representation on the monitor screen, but of all of the underlying details as well. So, I would expect 'no' to be the answer here. But, I've been wrong before.

[Dav]

Well...its not the monitor that controls how the graphics gets generated, it's the computer. The difference between a raster output and a vector output is not just one of representation on the monitor screen, but of all of the underlying details as well. So, I would expect 'no' to be the answer here. But, I've been wrong before.

1018194[/snapback]

 

 

This has been discussed many times on RGVAC, you can search there for technical details. The answer is no for any reasonable amount of money. It's theoretically possible to make a device that stores the vectors in ram and outputs them a video screen. It would be cheaper to build a new color vector monitor. And the results would look a whole lot better.

[EricBall]

You are confusing two different topics.

 

The ZVG allows a customized version of MAME to emulate vector games using a vector monitor (color, B&W, or even a Vectrex).

 

But when it comes to the actual original vector arcade game PCBs, you must use the appropriate vector monitor. And since replacements aren't being made anymore, nor is it economically feasible to build one from scratch, we're stuck with canibalizing one vector game to bring another back to life. (With the irony that sometimes a rarer game gets used as a donor for a more popular game.) Due to completely different technology there's no way an LCD could be used in place of a vector monitor.

 

The alternative is to use a hi-res PC monitor and MAME. Although it's not the same as a vector monitor (particularly in cases where the vector brightness is overdriven) it's usually much cheaper than a vector monitor.

[sebastius]

Actually, there is a way (for B/W games) i think. The site fpgaarcade.com has a bit of software for an fpga that converts a vector signal to SVGA! So I think it's possible, but difficult.

[supercat]

Actually, there is a way (for B/W games) i think. The site fpgaarcade.com has a bit of software for an fpga that converts a vector signal to SVGA! So I think it's possible, but difficult.

1020379[/snapback]

 

What do vector games output? A pair of +/- 5V signals for XY and a 0-5 volt signal for gating (three for RGB games) or something like that? Are the signals ramped going to the displays? How fast are they?

 

It shouldn't be too hard to design a board to do the conversion (I'd target a color LCD rather than a monitor, but that's just my taste). Basically, it would use a fast RAM as a framebuffer and run the following steps repeatedly:

 

-1- Clock out a pixel to the LCD/CRT.

 

-2- Do table lookups on those R,G,B values and store the result back to RAM. While doing this, sample X,Y,R,G,B from the analog inputs.

 

-3- Read the RAM at X,Y (call the values read r,g,b)

 

-4- Do table lookups with (R,r) (G,g) and (B,b) and store those values into RAM.

 

Generating a 60Hz 1024x768 display without antialiasing would require about a 200MHz memory bandwidth assuming an A/D sampling rate of 50MHz. Zippy, but not unreasonable (3MB of static RAM would not be unreasoanbly expensive).

 

Adding antialiasing while keeping the same clock rates would require splitting the memory into four sections (even rows/columns, even rows/odd columns, odd rows/even columns, and odd rows/columns). The board would start to get big (each section would need three 256x8 RAM chips and an FPGA) but shouldn't be too impractical.

 

Overall, I would expect the chip cost of the device to be not too far different from a high-end VGA card. Unfortunately, I doubt the volume on the thing would allow the engineering costs to be recouped at a reasonable price.

[+batari]

But when it comes to the actual original vector arcade game PCBs, you must use the appropriate vector monitor. And since replacements aren't being made anymore, nor is it economically feasible to build one from scratch, we're stuck with canibalizing one vector game to bring another back to life.

There is no functional difference between a so-called "vector CRT" and a "raster CRT." What makes a CRT vector or raster is the deflection yoke and deflection circuitry. The CRT itself doesn't care what sort of deflection it gets...

 

So it seems that someone somewhere would have found a suitable monochrome raster monitor and simply swapped yokes and adapted the CRT neck connectors, or at least tried to, so the CRT shortage would no longer exist? For that matter, I'd imagine that some CRT neck pinouts on vector and raster are the same, making things easier.

Edited February 19, 2006 by batari

[Dav]

But when it comes to the actual original vector arcade game PCBs, you must use the appropriate vector monitor. And since replacements aren't being made anymore, nor is it economically feasible to build one from scratch, we're stuck with canibalizing one vector game to bring another back to life.

There is no functional difference between a so-called "vector CRT" and a "raster CRT." What makes a CRT vector or raster is the deflection yoke and deflection circuitry. The CRT itself doesn't care what sort of deflection it gets...

 

So it seems that someone somewhere would have found a suitable monochrome raster monitor and simply swapped yokes and adapted the CRT neck connectors, or at least tried to, so the CRT shortage would no longer exist? For that matter, I'd imagine that some CRT neck pinouts on vector and raster are the same, making things easier.

1020688[/snapback]

 

I believe the only thing lacking for making new monitors is the yokes. There was someone who wound his own a while back, and was looking at making a run of montiors but that fell through I think.

[supercat]

There is no functional difference between a so-called "vector CRT" and a "raster CRT."  What makes a CRT vector or raster is the deflection yoke and deflection circuitry.  The CRT itself doesn't care what sort of deflection it gets...

1020688[/snapback]

 

Perhaps among older CRTs which used a curved screen and constant focus.

 

On newer CRTs, the vertical deflection, horizontal deflection, and focus all interact with each other. This allows screens to have shorter necks without turning into "fishbowls", but would make it difficult to achieve acceptable vector scanning performance.

[sebastius]

What do vector games output?  A pair of +/- 5V signals for XY and a 0-5 volt signal for gating (three for RGB games) or something like that?  Are the signals ramped going to the displays?  How fast are they?

 

It shouldn't be too hard to design a board to do the conversion (I'd target a color LCD rather than a monitor, but that's just my taste).  Basically, it would use a fast RAM as a framebuffer and run the following steps repeatedly:

 

-1- Clock out a pixel to the LCD/CRT.

 

-2- Do table lookups on those R,G,B values and store the result back to RAM.  While doing this, sample X,Y,R,G,B from the analog inputs.

 

-3- Read the RAM at X,Y (call the values read r,g,b)

 

-4- Do table lookups with (R,r) (G,g) and (B,b) and store those values into RAM.

 

Generating a 60Hz 1024x768 display without antialiasing would require about a 200MHz memory bandwidth assuming an A/D sampling rate of 50MHz.  Zippy, but not unreasonable (3MB of static RAM would not be unreasoanbly expensive).

 

Adding antialiasing while keeping the same clock rates would require splitting the memory into four sections (even rows/columns, even rows/odd columns, odd rows/even columns, and odd rows/columns).  The board would start to get big (each section would need three 256x8 RAM chips and an FPGA) but shouldn't be too impractical.

 

Overall, I would expect the chip cost of the device to be not too far different from a high-end VGA card.  Unfortunately, I doubt the volume on the thing would allow the engineering costs to be recouped at a reasonable price.

1020545[/snapback]

You are forgetting one small detail: A line 'degrades' on the fosfor after it's been drawn, the game needs to constantly update the screen and so you need to keep track of every vector, how long it's been there etc etc.

 

When in trouble, one can always use an oscilloscope as an example, that basically is a XY monitor with Z control.

[supercat]

-2- Do table lookups on those R,G,B values and store the result back to RAM.  While doing this, sample X,Y,R,G,B from the analog inputs.

You are forgetting one small detail: A line 'degrades' on the fosfor after it's been drawn, the game needs to constantly update the screen and so you need to keep track of every vector, how long it's been there etc etc.

 

That's what step 2 is for.

 

When in trouble, one can always use an oscilloscope as an example, that basically is a XY monitor with Z control.

1020771[/snapback]

 

Yeah, but I don't know of any "real" color scopes.

[TheBman80]

I believe the only thing lacking for making new monitors is the yokes.  There was someone who wound his own a while back, and was looking at making a run of montiors but that fell through I think.

1020700[/snapback]

 

 

I wish I could get new yokes for them my problem might be over I discovered a sight that sells replacement tubes for the 19VARP4 (B&W vector), 19VLUP22 (color vector), so if I could just get yokes for the tubes I could buy boards off eBay. It would greatly reduce the cost but finding someone that has yokes is such a pain I never see them go up often on eBay if at all. So if anyone knows anyone who makes or has any yokes for these tubes or even has an XY board set there willing to sell please let me know.

Edited February 21, 2006 by TheBman80

[TheBman80]

Well now I did acquire an entire 19V2000 board set Yoke and everything off an eBay auction said they had some problem with the Amplifier circuit because a red light was blinking on the board. Hopefully I can fix this would anyone know what the problem with this might be. I haven't yet received the boards but I want to know what I might be getting in to. It would be great if there was a place that could make these yokes now that I have one I could send it out and get someone to make them.

 

Edited March 1, 2006 by TheBman80

[+batari]

Well now I did acquire an entire 19V2000 board set Yoke and everything off an eBay auction said they had some problem with the Amplifier circuit because a red light was blinking on the board. Hopefully I can fix this would anyone know what the problem with this might be. I haven't yet received the boards but I want to know what I might be getting in to. It would be great if there was a place that could make these yokes now that I have one I could send it out and get someone to make them.

 

         

1026264[/snapback]

Plastic can be molded, and there are lots of companies that specialize in rewinding electric motors, so maybe one of them could do the wire winding. I'm not sure if the vector CRT's have the annular magnets on them for convergence adjustment, but if so, these are available in various sizes.

 

What may be hard to find is the ferrite core. I think these need to be of a particular shape. Of course, you'd have to tear apart your current yoke to find out...

 

I could be missing something too...

[TheBman80]

Well it took a long time to get these parts off eBay but their finally in my hands an since I have them now I figured that I better share some pictures. For those who have not seen one off a monitor here’s a 19V2000 yoke all sides.

 

 

 

And here's a picture of one of the boards.